Distinguished Speaker Series: Structural modeling of cellular interactome

Edmond J. Safra Center for Bioinformatics

Distinguished Speaker Series Seminar

Prof. Ilya Vakser

Director, Computational Biology Program and Center for Computational Biology, Department of Molecular Biosciences, The University of Kansas

"Structural modeling of cellular interactome"

Wednesday, February 21 2024, at 16:15 

Sherman building, room 632, Life Sciences Faculty

Refreshments from 16:00

Abstract: Docking-based simulation of cellular environment at atomic resolution samples the intermolecular energy landscape mapped by docking, which focuses on large ensembles of predicted models/energy minima corresponding to transient interactions that dominate crowded environment inside cells. Our DOCKGROUND resource provides a collection of datasets for the development and testing of docking techniques. The interactive web interface allows search by various parameters, such as release date, multimeric state, complex type, structure resolution, etc., visualization of the search results with a number of customizable parameters, and provides downloadable datasets with pre-defined levels of sequence and structure redundancy. The crowded environment drastically changes protein recognition properties, and thus significantly alters the underlying energy landscape. We addressed the effects of crowding on the protein binding funnel, distribution and size of the energy basins, and energy function smoothing. A comprehensive analysis of protein diffusion in different crowding condition, based on the landscape sampled by Markov Chain Monte Carlo simulations, determined translational and rotation diffusion rates for different types of proteins in crowding conditions in a broad range of concentrations. Recent developments of the atomic resolution simulation protocols introduced sophisticated forcefields and drastically increased the simulation trajectory length. While our previously developed approach had been already ~1M times faster than the alternative techniques, the new parallelized protocols further speed it up by orders of magnitude, allowing us to address physiological mechanisms that have been far beyond capabilities of atom-resolution modeling.

Host: Prof. Nir Ben-Tal, Life Sciences Faculty